Hepatitis B virus (HBV), which causes liver cirrhosis and hepatocellular carcinoma, has a small, compact genome. The only transcript required for genome replication is the 3.5-kb pregenomic (pg) RNA. It serves as the dicistronic mRNA for the translation of both core and P proteins, and also as the genome precursor to be encapsidated and converted by the P protein to double stranded DNA. Since a capsid is assembled from 240 copies of core protein but packages probably just one molecule of P protein, efficient genome replication requires a proper ratio of core / P protein translation. P protein translation involves ribosomal leaky scanning of upstream AUG codons including those of the core gene, but the control mechanisms remain ill defined. Genotype G harbors a unique 36-nt insertion at the 5' end of its core gene, which markedly enhances core protein translation whether in genotype G or when artificially introduced to other genotypes. This increase in core protein translation is rather associated with impaired genome replication in non-G genotypes, most likely due to a corresponding reduction in P protein translation. Paradoxically, deleting the 36nt from genotype G also impaired genome replication. We propose that the unique structural features of genotype G provide a window of opportunity to elucidate the control mechanisms regulating the translation of core vs. P protein. Aim 1 of this R21 grant application will verify the hypothesis that the insertion creates a hairpin structure downstream of core gene AUG to augment translation initiation. Aim 2 will validate a small open reading frame upstream of the core gene (the uORF) as a positive regulator of P protein translation. Aim 3 will investigate why genotype G harbors two nonsense mutations in the precore region. Besides pg RNA, HBV produces another 3.5-kb RNA with about 30-nt extension at the 5' end. This precore (pc) RNA is devoted exclusively to the translation of precore/core protein, the precursor to hepatitis B e antigen (HBeAg). The late stage of chronic HBV infection often selects for the G1896A nonsense mutation in the precore region to abolish HBeAg expression. Curiously, genotype G harbors an extra C1817T nonsense mutation at codon 2. We propose that C1817T enables translational re-initiation at the uORF instead of the core gene, thus redirecting the pc RNA towards P protein translation and rescuing genotype G replication despite the 36-nt insertion. Our studies will clarify translational control of core and P protein expression, and help understand how different HBV genetic variants avoid deviating from an optimal core/P protein ratio required for efficient genome replication.